Trypanosoma cruzi, a protozoan parasite, is the causative agent of Chagas' disease (American trypanosomiasis), an infection of cardiac muscle cells and nerve ganglia that affects over 20 million persons in South and Central America, with over 90 million at risk. For practical purposes, chronic Chagas' disease is incurable, and drugs used for early infections can produce adverse side effects. Our proposed research builds upon preliminary results that demonstrate a parasite cysteine protease (cruzain) is essential for parasite replication and transformation between stages of the T. cruzi life cycle. One approach has been the synthesis of pseudopeptide inhibitors based upon the known substrate specificity of cruzain. Design of these inhibitors has been aided by our successful crystallization of the target enzyme and solving of the crystal structure at 2.3 angstroms. These pseudodipeptides have sidechains which maximize binding to the cruzain active site, while at the same time employing synthetic derivatives with non-natural amino acid homologues to maximize in vivo half life. A second modification has been to explore a number of different blocking groups to enhance solubility and therefore oral bioavailability. Finally, we have moved to new chemistry at the carboxy terminus to take advantage of the unique properties of the active site cysteine thiol group. These derivatives are inert to other thiol groups but which bind covalently to cysteine protease active site. A group of these compounds has now been tested both in in vitro culture assays to assess their effect on parasite replication, and in an in vivo murine model of infection. Two derivatives in which the sidechain amino acids are homophenylalanine and phenylalanine, were effective both in vitro and in vivo without evidence of toxicity to mammalian cells or tissues. The second approach we have taken is to identify leads utilizing computer searches of chemical databases using the crystal structure of cruzain as a template. One lead compound, a dihydrazide derivative, is similar to inhibitors we have also shown are effective against Leishmania and malaria cysteine proteases. This derivative was amplified through combinatorial chemistry, and several hundred modified analogues tested. A dimethylamine derivative of the original dihydrazide has been shown to arrest replication of T. cruzi in vitro without toxicity to mammalian cells. This compound has now also been shown to protect mice from a lethal infection. As with the dipeptide analogues, our next phase of work will include modifications to enhance aqueous solubility and oral bioavailability of the DOCK-derived compounds. We have studied the effects of protease inhibitors on the T. cruzi parasite itself. A surprising result was that the earliest detectable intracellular abnormality in the parasite is a distinct swelling of the Golgi compartment. This is followed by swelling of the endoplasmic reticulum and death of the parasite. Utilizing specific antibodies against cruzain, we have shown that the enzyme is arrested in its trafficking pathway at vesicles budding from the lateral Golgi. Noteworthy, there is no effect of inhibitors on Golgi of mammalian cells, even those in which amastigotes are residing. Our working hypothesis is that the inhibitors are preventing normal processing of the enzyme as an autocatalytic event, thereby retaining unprocessed enzyme in an early trafficking compartment. In the next phase of our work, we will test this hypothesis using mutants of cruzain in which the processing sites have been altered, and a transfection system to allow for overexpression of these mutants.